This post was originally published as a press release on 2 September 2021.

A new study has filled in details in the picture of online misogyny in Singapore.

From April to August 2021, gender equality organisation AWARE and technology firm Quilt.AI conducted a collaborative research project using a proprietary machine-learning model to analyse publicly available online data. Researchers employed this model to study online discourse around gender-based violence, as well as opposition narratives to gender equality. The study focused on Singapore-based online activity, and charted how the above were impacted by COVID-19.

“We were excited to partner with the innovative team at Quilt.AI on this overdue look at online harms in the local context,” said Shailey Hingorani, AWARE’s Head of Research and Advocacy. “Our findings can inform interventions to address the scourge of misogyny as it gains ground in society, and make the online experiences of women and girls safer.”

Discourse on Gender-Based Violence

Quilt.AI’s machine-learning model was used to examine the online discourse around gender-based violence (GBV) in Singapore, as well as its volume, platforms and main proponents. The research team studied 700 public posts across Facebook, Instagram, Twitter, Reddit and HardwareZone; as well as 1,620 comments on YouTube relating to GBV. From here, three main categories of “myths” (i.e. fallacies used to defend and justify GBV) were observed: confusion around consent (40%), diminishing or dismissing GBV (35%) and other forms of victim-blaming (25%). The model did also pick up on posts that featured outrage at and critique of these myths, as well as frustration towards institutions for not having better prevention and response systems for sexual violence.

The research team also studied 819,190 unique searches (between March 2019 and March 2021) of keywords relating to gender-based violence in Singapore. Searches relating to rape, sexual assault and consent all increased in the past year. On the other hand, searches about physical violence (specifically, intimate partner violence) and relevant help-seeking peaked around Singapore’s first COVID-19 circuit-breaker period (March-May 2020).

Opposition Narratives to Gender Equality

Researchers also identified five opposition narratives to gender equality from a sample of 500 posts on Twitter, Facebook and HardwareZone, using two parameters—frequency of posts and degrading misogynistic language aimed at women and girls—to classify each narrative as “extreme”, “moderate” or “weak”. The most prominent opposition narrative was found to be resentment over National Service (NS): Men bring up mandatory NS as a rebuttal in myriad conversations about inequalities and oppressions faced by women. Other prominent narratives centred on misconceptions around the aims of feminism, with men stating that the movement calls for “female dominance” as opposed to equality, and identifying perceived hypocrisy tied to imagined expectations of “gentlemanly” behaviour such as paying for dates.

High Engagement Rates

Lastly, the model found that female accounts on Twitter received twice as many misogynistic comments as a random sample of accounts. In terms of engagement, misogynistic comments were twice as likely to be “liked” and 4.5 times more likely to be retweeted when compared to non-misogynistic comments.

“These findings confirm what we have long suspected that women were experiencing online in Singapore, based on anecdotal evidence from AWARE’s clients and community,” said Ms Hingorani. “The speed at which misogynistic ideas proliferate online—when generated and promoted by anonymous users with little accountability for the harm they cause—should deeply concern anyone with an interest in gender equality. We hope that tech companies, the government and civil society can work harder to stamp out online misogyny through greater education, initiatives to encourage bystander intervention, and a duty of care to moderate user-generated data and images.”

“As a mission based technology firm, gender and climate are our two areas of focus,” said Anurag Banerjee, CEO of Quilt.AI. “A large part of the gender work done by us focuses on fighting misogyny online. We are glad to have been able to have supported AWARE’s efforts.”

Read the full report, “A study of online misogyny in Singapore”, here.

About AWARE

AWARE is Singapore’s leading women’s rights and gender-equality advocacy group. It works to identify and eliminate gender-based barriers through research, advocacy, education, training and support services. AWARE embraces diversity, respects the individual and the choices she makes in life, and supports her when needed. aware.org.sg

About Quilt.AI

Quilt.AI is a consumer research and behavior change company based in Singapore and the USA. It works with more than 100 organisations globally on increasing awareness and knowledge of human behavior on the Internet. Quilt.AI has been rated by the Economist as an “AI for Good” company as it focuses on using technology to address inequalities in gender and climate globally. quilt.ai